Tea Totaling: Green Tea and Cholesterol

Abstract & Commentary

By Russell H. Greenfield, MD

Synopsis: A well-done meta-analysis showed that green tea, either as a beverage or as a supplement, could help lower total and LDL cholesterol levels in a statistically significant manner, but has no effect on HDL. The degree of clinical impact is debatable; what is not debatable is the poor quality of most studies on green tea and cholesterol.

Source: Zheng X-X, et al. Green tea intake lowers fasting serum total and LDL cholesterol in adults: A meta-analysis of 14 randomized controlled trials. Am J Clin Nutr 2011;94:601-610.

The authors performed a meta-analysis of randomized, controlled trials to ascertain the effect of regular green tea ingestion, in the form of tea or green tea extract, on total cholesterol (TC), LDL cholesterol, and HDL cholesterol in adults. A comprehensive, systematic literature search of English-language reports of clinical trials was performed. Articles were located in PubMed (1967-2010), Embase (1977-2010), and the Cochrane Library database, and from reviews and reference lists of relevant articles. The authors state that attempts also were made to contact investigators for access to unpublished data.

Inclusion criteria were strict; only studies in which subjects consumed green tea or its extract for > 2 weeks, were randomized and of a parallel or crossover design, where pre- and post-intervention lipid levels were retrievable, where food intakes between active and control groups were similar, where a supplement was not given as but one component of a multi-pronged supplement, and where fasting blood samples had been obtained were considered. Study quality was assessed using Jadad score, and publication bias was determined using funnel plots and the Egger's regression test. Search, data extraction, and quality assessment duties were completed independently by two reviewers, and discrepancies resolved through discussion (presumably amicably). Study characteristics (including sample size, study design, study duration, dose, and type of intervention), population information, baseline and final concentrations or net changes of TC, LDL, and HDL were extracted.

A total of 805 potentially eligible articles were initially identified, 778 of which subsequently were excluded either because they were not clinical trials or the interventions were not relevant to the meta-analysis. Thus, 27 articles were selected for detailed evaluation, 13 of which were excluded due to a variety of methodological shortcomings including short trial duration and lack of relevant outcomes. In the end, 14 randomized controlled trials providing data from 1136 subjects were analyzed. Trial size ranged from 20-240 subjects, duration from 3 weeks to 3 months (median 12 weeks), and dose of green tea catechins from 150 to 2500 mg/d (median: 625 mg/d). Five trials were conducted in healthy adults with another five focusing primarily on overweight and obese adults. A green tea beverage was tested in half of the trials, an extract in the remaining half. Twelve trials used a parallel group study design, two employed a crossover design. Twelve studies were double-blinded, and 10 were placebo-controlled (how double-blinding was maintained using a green tea beverage was not explained). In most of the trials investigators attempted to maintain participants' usual lifestyles, but in two studies a low-fat diet was used, and a low-energy diet in another.

Study quality was very disappointing, with only three of 14 studies assessed as being of high quality based on Jadad score. Two studies received industry funding.

Primary outcome measures of interest were pre- and post-intervention changes in TC, LDL, and HDL. Results for TC were reported in 14 comparisons from 13 studies (n = 949). Mean change in TC concentration was significantly reduced in subjects receiving green tea compared with controls (-7.20 mg/dL; 95% confidence interval [CI], -8.19 to -6.21 mg/dL; P < 0.001). Mean change in LDL concentration was reported in 11 comparisons from 10 studies (n = 853) and was significantly decreased by 2.19 mg/dL in the intervention groups compared with controls (95% CI, -3.16 to -1.21 mg/dL; P < 0.001). HDL results were calculated in 12 comparisons from 11 studies (n = 998), and in the intervention groups a favorable trend was identified that was not statistically significant (+ 0.25 mg/dL; 95% CI, -0.73 to 1.23 mg/dL; P = 0.62).

Subgroup analyses were conducted to explore methodological impacts such as dose-effect relationships and differences between drinking green tea and the taking of green tea extracts. Catechins intake was divided into low dose (< 625 mg/d) and high dose (> 625 mg/d). Time to follow-up varied from 3 weeks to 3 months; a subgroup analysis was performed by dividing the follow-up duration into a shorter-term subgroup (< 12 wk) and a longer-term subgroup (> 12 wk). These analyses showed that the reported reductions in TC and LDL were not influenced by type of intervention (drinking green tea or taking a supplement), and that TC and LDL were significantly decreased in both lower- and higher-catechin consumption groups. Green tea significantly reduced TC and LDL in both healthy subjects and in participants with cardiovascular risk factors. Short- and longer-term intervention subgroups experienced significant reductions in TC and LDL cholesterol.

Neither study heterogeneity nor publication bias was identified. Removal of the two trials associated with industry funding did not change the final results. The authors did not pool data on safety, stating that no serious side effects were reported in the trials they examined.

The authors concluded that the regular ingestion of green tea, whether through drinking or the taking of supplements, results in significant reductions in TC and LDL concentrations, but no significant effect on HDL.


Cardiovascular disease is a multifaceted disorder, but the contribution of dyslipidemia to increased risk of heart attack and stroke is universally accepted. Diet and exercise remain the mainstays of therapy, often are very successful, and may help to negate the need for medical therapy or at least lessen the dosage required to attain healthier cholesterol levels. Any easy, widely available intervention that could add to the success of dietary means of lowering cholesterol would be welcome, especially in the form of a beverage. Therein lies the value of this meta-analysis.

Green tea (Camellia sinensis) is one of the most widely consumed beverages in the world and is viewed by many as a health-promoting powerhouse with demonstrated antioxidant, anticancer, anti-inflammatory, and antithrombotic activity. Animal data suggest it also may help lower cholesterol levels. The majority of human trials also suggest a lipid-lowering effect of regular green tea consumption, but not all, and that was the stated reason for the researchers to embark on this meta-analysis: to try and make sense of the available human data on green tea and its potential cholesterol-lowering action. And they did, but only to a point, with the shortcomings occurring through no fault of their own.

The authors suggest that the positive results they report point to a potentially significant role for green tea as part of a public health dietary policy to help improve cardiovascular health. This is over-stated. Green tea indeed may offer health benefits, including beneficial effects on lipids, but the current study does not add in a compelling way to the discussion. Even though study quality did not affect the results of meta-analysis, the majority of reviewed trials were of poor quality and short duration. In addition, the researchers rightly note that no definitive recommendation regarding proper dose of green tea catechins to promote cardiovascular health can be drawn from their investigation due to the wide range of doses employed across studies. However, it is reassuring to know that the positive effects of green tea on TC and LDL were consistent across varied methodologies. Then again, some might argue that the impact of green tea on cholesterol levels reported here may not be clinically relevant.

This meta-analysis is somewhat helpful in that it suggests green tea may hold promise as a mildly effective adjunct in the management of hyperlipidemia. Perhaps more importantly, it serves notice on the poor state of research regarding that promise. The good news is that practitioners should generally feel comfortable recommending green tea to their patients with high cholesterol, especially as a beverage, considering both safety and clinical utility. Optimal dosing has yet to be agreed upon, but most authorities suggest 3-5 cups (1-2 mugs) daily. Questions regarding the regular use of green tea extracts persist, however, in light of reports suggesting the potential for hepatotoxicity. The tea tastes way better, anyway.